New Frailty Index Approach Predicts COVID-19 Mortality Risk

Abstract The relationships between blood biomarkers, frailty, and the risk of death of people diagnosed with COVID-19 is unclear. In the current investigation we decided to analyze the collective effect of multiple biomarkers (laboratory markers of inflammation, blood biochemistry deviations, comorbidity, demographics) on mortality in people diagnosed with COVID-19. We analyzed baseline data of one hundred fifty-five patients (age range from twenty-six to ninety-four) diagnosed with COVID-19. Thirty-seven parameters (including major morbidities) were used to derive the frailty index (FI) and calculate the risk of death as a function of FI and individual biomarkers. Discriminative ability was assessed by the area under the receiver-operating characteristic (ROC curves). The mean frailty index was 0.17 (SD = 0.10), FI of those who survived was 0.11 (SD = 0.078) and those who died was 0.22 (SD = 0.093). In a sex-adjusted model, the FI was a more powerful predictor for mortality than age. The ROC analysis showed that models involving FI as a feature have good discriminative ability for predicting COVID-19 mortality: AUC for age was 0.77, for the FI it was 0.82, and for the fully adjusted model (age + FI) it was 0.84. Thus, the systemic effect of multiple biological processes comprising aging are elucidated using the Frailty Index approach. Assessment of the frailty index at the time of admission of a patient with COVID-19 to the clinic can help to predict the high risks of severe disease and mortality.